 The soluble carrier hormone binding protein, HBP, is an important factor in the growth of humans and other animals. It can bind to hormones and regulate their activity. However, identifying HBP is difficult due to its complex structure and limited availability of data. To address this issue, researchers have developed a machine learning-based approach to identify HBP. This method encodes samples using optimal tripeptides derived from the binomial distribution. The resulting model achieved an overall accuracy of 97.15% in a five-fold cross-validation test. A user-friendly web server, HBPRED 2.0, has been created to facilitate access to the model. This article was authored by Joe Xiantan, Shirhowli, Zimei Zhang, and others.